Technical Abstract: Rangeland ecosystems are often characterized by a patchy mosaic of vegetation types, making measurement and modeling of surface energy fluxes particularly challenging. The purpose of this study was to evaluate surface energy fluxes measured using eddy correlation at three rangeland sites, and use the data to improve simulations of turbulent energy fluxes in a multi-layer plant canopy model, the Simultaneous Heat and Water (SHAW) model. Model modifications included adjustment of the wind profile roughness parameters for sparse canopies, and introducing Lagrangian far field turbulent transfer equations (L-theory) in lieu of the K-theory approach used previously. There was relatively little difference in simulated energy fluxes for the aspen canopy using L-theory versus K-theory turbulent transfer equations, but L-theory tracked canopy air temperature profiles better. Upward sensible heat flux was observed above aspen trees, within the aspen understory, and above sagebrush throughout the active snowmelt season. Model simulations confirmed the observed upward sensible flux during snowmelt was due to solar heating of the aspen limbs and sagebrush. Thus, the three EC systems were unable to properly quantify fluxes at the snow surface when vegetation was present, illustrating the difficulty in using EC systems to quantify energy fluxes at the snow surface with vegetation present. Good agreement between measured and modeled energy fluxes suggest that they can be measured and simulated reliably in these complex environments, but care must be used in the interpretation of the results.